33074 research outputs found
Sort by
Automated autoencoder neural network architecture design
Full text linkMaster of Science in Computer Science, North-West University, Potchefstroom CampusWithin recent years, there has been a sharp rise in the volume of raw data produced, analysed, and utilised across various fields. With the rapid evolution of technology, information systems, and data acquisition methodologies, this exponential growth necessitates advanced systems capable of efficiently processing, analysing, and interpreting large datasets. Autoencoders, a class of neural networks, have demonstrated significant potential in applications, such as dimensionality reduction, feature extraction, anomaly detection, thereby making them powerful tools for managing vast amounts of data. However, the manual design and development of autoencoder architectures are complex, time-consuming, and resource-intensive tasks that require expert knowledge, often leading to suboptimal models. The primary aim of this study is to develop an automated algorithm for constructing accurate autoencoder neural network architectures. The proposed Automated Autoencoder Neural Network Architecture Design (AANNAD) algorithm automates the hyperparameter selection process, iteratively generating and improving autoencoder models until a predefined resource limit is reached. This study follows a positivistic research paradigm and employs a Design Science Research methodology to evaluate the performance of the AANNAD algorithm across three datasets consisting of different data types originating from various application areas. These include the MNIST, SOCOFing, and credit card fraud detection datasets. The algorithm's performance is measured against traditional methods using reconstruction error metrics. The results demonstrate that the AANNAD algorithm consistently produces high-performing autoencoder architectures, often surpassing models specifically designed for their respective tasks. These findings suggest that the AANNAD algorithm may significantly streamline the development process of deep learning models, reducing time and human effort while maintaining or improving accuracy.Master
Unveiling Multicultural and Multilingual Positioning Theory as Normative Ideal in Multilingual Contexts
Full text linkJournal Article, Faculty of Humanities, International Journal of Indigenous Language Media and DiscourseRapid and commendable change has taken place in many post-colonial societies. The past few years have been epochal in transforming traumatised and fragmented societies into more stable and democratic ones. This was achieved through various means of political accommodation, including recognising the multiple cultures of those polities. Yet, despite the fanfare about multiculturalism accommodation, linguistic diversity and the quality of its actual practice remain parlous. Realising this problem, a new Multicultural and Multilingual Positioning Theory is proposed and developed to foster integration, focusing on the recognition of indigenous languages and their usage in various public platforms. This article highlights pertinent issues inherent in multilingual contexts using South Africa as an example. Insights from Multiculturalism Accommodation are used as a starting point to argue that the perilous situation in South Africa not only limits indigenous language communities' effective participation in the national life but also constrains citizens' constitutional rights. The article argues that although the country recognised nine African languages to the status of official languages, the actual usage of such languages in official discourse remains very erratic and effete Ultimately it proposes a novel theory, the Multicultural and Multilingual Positioning theory, as an alternative approach to ensure that multiple languages and cultures function effectively in multicultural and multilingual contexts. This theory contributes to new epistemic horizons and provides a framework for practical implementation
Communication, Media, Culture, and Evolving Trends in the World of Indigenous Languages
Full text linkIndigenous language is a significant component of the cultural heritage of every community or nation. Aside from serving as a medium of communication, language serves as a window into a community or nation’s historical and cultural heritage. Unfortunately, in the past century, many indigenous languages have suffered threats to their survival and continued use. This development is due mainly to colonialism and the wave of globalization that started on the eve of the 20th century. The so-called “world languages,” such as English, French, Spanish, German, and Chinese are taking over as more preferred and widely spoken in indigenous communities despite their foreign origin and the long history of indigenous languages in these communities. Unless effective revitalization programs are embarked upon, the loss of more indigenous languages is imminent. The positive news, however, is that certain trends have evolved in the application of indigenous languages which portend opportunities for the revitalization and flourishing of indigenous languages in this present age of globalization. This lecture examines the connections between indigenous languages, communications, media and globalization, as well as evolving trends and future trajectories in the application of indigenous languages, with a view to advocating for more deliberate steps in the preservation of indigenous languages by respective stakeholders
A sociolinguistic investigation into the South African Indian speakers of Mohadin and Potchefstroom
Full text linkMaster of Arts in English, North-West University-- Potchefstroom CampusThis dissertation investigates the linguistic variation within the South African Indian English (SAIE) community of Potchefstroom, and Mohadin, focusing on two key phonetic variables: retroflexion and the GOOSE vowel. Situated in a post-apartheid context, this study examines how inter-ethnolinguistic mobility, social class, age, and gender influence these phonetic features, particularly comparing speakers from the town of Potchefstroom and the historically segregated township of Mohadin. The study involved a total of 32 participants, divided equally between the two areas, and further segmented by age and gender. Utilizing a quantitative acoustic approach, data were collected through standard Labovian sociolinguistic interviews and analyzed using PRAAT and RStudio, focusing on Voice Onset Time (VOT) for /t/ and /d/ and second-formant frequencies for the GOOSE vowel, with statistical analysis performed through Welch’s t-tests and conditional inference trees (ctrees), to explore the effects of social variables on linguistic variation. These methods were selected due to their robustness in handling specific data characteristics, ensuring reliable and valid statistical inferences. Welch’s t-test is an adaptation of Student’s t-test designed for comparing the means of two independent groups when the assumption of equal variances is violated. Unlike Student’s t-test, which assumes homogeneity of variance, Welch’s t-test adjusts for differences in variance and sample size, making it more reliable under conditions where variance equality cannot be ensured (Ruxton, 2006). Given that linguistic variables such as Voice Onset Time (VOT) for /t/ and /d/, as well as second-formant frequencies for the GOOSE vowel, may exhibit different levels of variability across social groups (e.g., age, gender, and geographic location), Welch’s t-test provides an appropriate statistical tool for comparing group means while mitigating the risk of Type I errors. This aligns with the recommendation that Welch’s t-test should be the preferred approach in most cases of mean comparison due to its robustness and improved accuracy over Student’s t-test (Delacre, Lakens, & Leys, 2017). In addition to Welch’s t-tests, this study employed conditional inference trees (ctrees) to further explore patterns of linguistic variation. Conditional inference trees are a non-parametric statistical method used to analyze the relationship between predictor variables and an outcome variable in a decision-tree format. Unlike traditional regression models, ctrees do not assume linearity or normality, making them particularly useful for linguistic data, where interactions between social factors and phonetic features may be complex and nonlinear (Tagliamonte & Baayen, 2012). According to the Comprehensive R Archive Network (Hothorn, Hornik, & Zeileis, 2006), ctrees operate through a recursive partitioning process that avoids overfitting by applying statistical significance tests at each node of the tree, making them a more interpretable and robust alternative to traditional classification or regression trees (CART). Given the exploratory nature of this study, examining how age, gender, and geographic location influence phonetic variation—ctrees provide a valuable means of identifying interaction effects and patterns in the data that may not be readily apparent through traditional regression approaches. By incorporating both Welch’s t-tests and conditional inference trees, this study ensures a comprehensive statistical approach that is both theoretically justified and methodologically rigorous. Welch’s t-tests allow for precise mean comparisons across social groups while accounting for variance heterogeneity, whereas ctrees facilitate an in-depth, non-parametric exploration of complex social-linguistic interactions. The use of these methods aligns with the best practices in quantitative sociophonetics and contributes to the reliability and validity of the findings. The study reveals a significant variation in the use of retroflexion and GOOSE-Fronting, with younger speakers and women, particularly in the town of Potchefstroom, leading the linguistic shift towards more standardized English pronunciations. In contrast, older speakers, specifically women in Mohadin retain more traditional speech patterns, with greater use of retroflexion in comparison to the men. The analysis also highlights the complex interplay between social mobility, cultural norms, and language change within this unique community. The findings contribute to the broader understanding of language variation and change in postcolonial Englishes, offering new insights into how social factors such as gender roles, geographic location, and age shape phonetic variation in multilingual contexts. This study extends existing research on South African Indian English by focusing on an under-researched community, thereby providing a valuable addition to the field of sociolinguistics.-North-West University, (Faculty of Humanities)
-DSACMaster
Surface extended catalyst support structures for hydrogen energy applications
Full text linkThesis (Ph.D. (Chemical Engineering)) -- North-West University, 2025In the current study, metal foams are evaluated as process-intensifying, surface extended catalyst support structures for three hydrogen (H2)-based applications, namely the dehydrogenation of perhydro-dibenzyltoluene (H18DBT), the passive autocatalytic recombination (PAR) of H2 and the direct-CO2-to-methanol (MeOH) synthesis. For these applications, suitable in-house developed or commercial catalysts were washcoated on the surface of metal foams. They were investigated during catalytic activity tests using methods selected from the available literature. Their performances were then compared experimentally or using computational fluid dynamics (CFD) simulation to those of typical catalyst systems used in reactors conventionally used for the same processes. The dehydrogenation of H18DBT was conducted in an unstirred tank rector and a fixed-bed reactor (FBR). An in-house developed Pt/Al2O3 catalyst was coated on aluminium (Al) foams and tested in the tank reactor. The performance of the foam-supported catalyst was later compared to that of the same amount of Pt/Al2O3 pellets. Experiments were conducted at 300°C, and Pt/H18DBT mol. ratio of 0.05–0.4 mol.%. H2 productivities were 12–59% higher in the foam-based than the pellet-based tank reactor. The higher performance of the foam-based reactor was attributed to the geometric properties of the coated foam. The Pt/Al2O3-coated foams also demonstrated a high catalytic stability for more than 16 h on stream in a FBR. The relatively low degree of dehydrogenation obtained in a single pass in the FBR was improved by recycling the partially dehydrogenated product three times in the reactor at the same temperature and flow conditions. However, the H2 productivity significantly decreased with the number of dehydrogenation cycles. The PAR tests were performed in a recombiner consisting of Pt/Al2O3-coated Al foams in the catalytic section. The feed gas, made of 2–6% H2 in air, was introduced to the recombiner at inlet velocities of 2–10 m/s. H2 conversions between 57 and 89% were achieved in the recombiner, while the highest temperature recorded during the operations was 172°C lower than the self-ignition limit of a H2 air mixture. A comparison between the foam-based, plate-based and pellet-based recombiners using the same amount of catalyst revealed that the H2 conversions achieved on the coated foams were similar to those obtained on the coated plates. The pellet-based recombiner exhibited the lowest performance of the three systems. Non-uniform temperature profiles were observed on the coated foams/plates due to the unpredictable, uneven gas flow distribution in the recombiner. Three-dimensional (3D) heterogeneous CFD simulations of the foam-based recombiner were successfully validated using experimental data, including H2 conversions and maximum temperatures. The model can successfully be used as a numerical tool for the prediction of the recombiner performance at various conditions. The direct-CO2-to-MeOH synthesis was performed in a two-module, bench-scale FBR containing a commercial Cu/ZnO/Al2O3 coated on copper foams. Activity tests were conducted at weight hourly space velocities of 1.125–2.925 NL/gcat/h, pressures of 30 and 50 bar and temperatures of 190–250°C. CO2 conversions, MeOH selectivities and MeOH productivities up to 27.46%, 82.97% and 30 gMeOH/h, respectively, were achieved in the reactor. This demonstrated the suitability of the coated foam for the direct CO2 hydrogenation to MeOH. Temperatures measured at different points in the reactor revealed that the nearly isothermal conditions were achieved in the reactor. The experimental results of the direct-CO2-to-MeOH synthesis were used to validate a 3D pseudo-homogeneous CFD model of the foam-based modular reactor. The model included simplified and optimised kinetic expressions which were successfully used to predict the CO2 conversions, MeOH selectivity, MeOH weight-time yield (WTY) and CO outlet concentration. The 3D model was also used as prediction tool to simulate a modular packed bed reactor (PBR). Based only on the difference in geometric properties between the two reactors, the model predicted lower CO2 conversions, MeOH selectivities and MeOH WTY in the PBR compared to the foam-based reactor. The performance of the coated foams varied according to the application used. This study demonstrated that the selection of a foam-based catalytic system can be motivated by its ability to enhance processes which are more strongly affected by mass transfer than kinetic limitations
Unveiling the Hidden Risks
Full text linkShort-story, Faculty of Health Sciences (Nursing)-- North-West University, Mahikeng CampusIn the rural expanse of South Africa's Eastern Cape, where healthcare resources are often limited, a silent epidemic is unfolding. People living with HIV (PLWHIV), who have long been the focus of antiretroviral therapy (ART) efforts, are now facing an additional threat: type 2 diabetes mellitus (T2DM). A recent study conducted in the OR Tambo district has shed light on the lifestyle determinants contributing to this emerging crisis
Financial research statistics
No full texttbdfuvx kfdhvfhnhf fbjfbn nkfnb cj ksdbfcj fv ujsjshfbdsd j niuh sdosidh ksbfjhbv ukfxc cxjhmtghjhbsdsd c nksdjcbsdhlassddshbc kjrnfsdncldnewnbf kjhrefmscniasjhsd cdf kj ermcdsxksHuman Sciences Research Counci
Photosynthetic responses of Lessertia frutescens to different shading conditions
Full text linkMaster of Science in Botany, North-West University, Potchefstroom CampusLight manipulation is crucial for optimizing plant growth and photosynthetic performance. This study investigated how different shading conditions influence the photosynthetic efficiency of Lessertia frutescens, a medicinal plant with significant pharmacological value. Plants were grown under grey, white, and green (40%) laminated shade netting and compared to an unshaded control. Key photosynthetic parameters, including chlorophyll fluorescence, PSI redox states, absorption fluxes, cross-section parameters, and chlorophyll content, were measured using a Handy-PEA fluorescence meter and a CCM-300 chlorophyll content meter. Results showed that shading improved photosynthetic performance relative to full sunlight, likely due to reduced photoinhibition and optimized light interception. Shaded plants exhibited a higher maximum quantum yield (FV/FM) and total performance index (PITOTAL) values, lower minimal fluorescence (F0), and increased PSI oxidation and re-reduction rates, suggesting enhanced photoprotection. While green laminated shade netting was expected to provide the greatest benefit, no significant (p < 0.05) advantage was observed among shading treatments, indicating that overall shading, rather than spectral composition, played a dominant role. Intraspecific variation was evident, with plant group A responding more favourably to shading, while group B exhibited a decline in PSI efficiency with prolonged exposure to laminated shade netting. Chlorophyll content was higher in shaded plants, supporting improved light capture. The increase of the absorption flux per cross-section (ABS/CSm) confirmed efficient electron transport under shading, while unshaded plants showed increased energy dissipation (DI/CSm) as a photoprotective response. Overall, these findings demonstrate that moderate shading enhances photosynthetic efficiency in Lessertia frutescens and improves plant vitality. These insights contribute to refining shading strategies for improved cultivation and stress resilience.Master
The protection of the rights of child offenders in pursuit of juvenile justice in selected African countries
Full text linkMaster of Laws with International Aspects of Law, North-West University, Potchefstroom CampusChildren are among the most vulnerable populations in the world. Their vulnerability worsens when they are exposed to the law. Children in conflict with the law face numerous challenges and violation of their fundamental rights. The protection of their rights is a cause for concern and requires urgent attention from states, policymaker and social and civil organisations. In their quest to ensure that the rights of children are protected, the United Nations and African Union enacted several legal instruments. These legal instruments were enacted as a measure to preserve the rights of children. International standards provide a critical framework guiding national practices. These laws are binding to member states who are signatories to the legal instruments and who have ratified the legal instruments in their distinct jurisdictions. Previous studies that have been conducted have revealed that, although member states are signatory to the United Nations and African Union legislations and have made attempts to ratify these legal instruments, to date some of these member states are not effectively implementing these legal instruments. Consequently, the purpose of signing and ratifying these international law instruments is not being fully realised. Research has also proven that in some selected African countries, children in conflict with the law are ill-informed about their rights and end up being exploited. This study was therefore an attempt to bring to light a few of international laws that deal with the rights of child offenders, their purposes and aims. Findings of a comparative analysis of four selected African countries namely: South Africa, Kenya, Zambia and Zimbabwe are a confirmation of the gap that exists between enactment of legislation and implementation thereof. While these countries should be applauded for enacting legislation to protect the rights of children, there is evidence that they fall short of implementing the legislation thus failing to realise the purpose of the legislation. As a result, there is need to adopt best practice that resonates with the needs of children in conflict with the law.Master
Refining the methods for determining soil organic carbon stocks in South Africa
Full text linkDoctor of Philosophy in Science with Environmental Sciences, North-West University, Potchefstroom CampusThe aim of this research was to refine the methods for determining soil organic carbon stocks (SOCS) in South Africa. There were a few gaps that this study tried to address that the Verified Carbon Standard methodology, and the Gold Standard methodology are not being clear on or specific about. These gaps have a direct impact on the South African industry and includes: (i) the uncertainty regarding which analytical method should be used for determining soil organic carbon (SOC) content in South Africa, (ii) the unavailability of a near-infrared (NIR) calibration algorithm for SOC content in South Africa, (iii) the uncertainty on how to map SOCS at field scale in South Africa, and (iv) the uncertainty of using either a modelling or re-measuring approach be for quantifying SOCS at field scale in South Africa. Key objectives of this study included to: (i) evaluate the preferred method to be used for determining SOC content and the application for South Africa conditions, (ii) develop a NIR calibration algorithm for SOC content in South Africa on three different scales, (iii) determine how to map SOCS specifically at field scale in South Africa, and finally (iv) compare SOCS quantification approaches specifically at field scale in South Africa. The following paragraphs indicate the methodology for achieving the objectives, as well as the main findings and suggestions for the way forward. Accurate quantification of SOC content is essential for the assessment of carbon credits. In South Africa, the standard methodologies for carbon credit assessment does not specify which analytical method should be used for determining SOC content. The study aimed to determine which analytical method should be used for determining SOC content for the assessment of carbon credits. Secondly, it determined whether pedotransfer functions could be used for transferring SOC content values between methods, especially for transferring historic determined SOC content to more recently determined values using more modern methods. Two-hundred-and-twenty topsoil (0–30 cm) samples were collected from five catchments in South Africa and analysed for SOC content with the three analytical methods: Walkley-Black wet-oxidation (WB), total dry combustion (TDC) and loss-on-ignition (LOI). The study found that the TDC method provided the most reliable SOC content values and should still be considered the preferred method for determining SOC content for the assessment of carbon credits in South Africa. The WB method should be avoided if a soil is expected to have a high SOC content, while the LOI method could still be used for determining SOM, however, this method should be avoided when determining SOC content. The study also reached the second aim by successfully creating pedotransfer functions between all three methods. However, only the WB and TDC methods had a very strong relationship (R² = 0.91) and showed that accuracy start to decrease significantly above 2.5% SOC content. Therefore, the pedotransfer function (SOCWB = -0.157 + 0.895 x SOCTDC – 0.0149 x SOCTDC² – 0.000606 x SOCTDC³) could be used for transferring SOC content values with SOC content up to 2.5%. Near-infrared (NIR) spectroscopy has emerged as an easy, rapid and cost-effective alternative for SOC analysis and the accounting of carbon credits. South Africa currently lacks a calibration algorithm for predicting SOC content from NIR spectroscopy. This study aimed to develop a NIR spectroscopy calibration algorithm for SOC content, specific to South Africa. Soil samples were collected from 2 fields and 3 catchments across South Africa. These samples were analysed using the total dry combustion (TDC) method and scanned with a NIR spectrometer. Sixty NIR calibration algorithms were developed on a regional scale. The impact of methodological parameters, such as sample state, sampling design, processing and machine learning models, on the root mean square error (RMSE) of the validation statistics was also assessed. Although 60 regional-scale calibration algorithms were developed, none were suitable (RMSE = 0.39 % and RPIQ > 2) for SOC content prediction, which was attributed to the small sample size (n = 238). However, local calibration models for the Tsitsa catchment and Ottosdal fields presented great accuracy (RMSE 1.5) that can be used for future SOC content prediction. The study found that the open spectral library global prediction model poorly predicted SOC content using local data (RMSE = 1.23 % and R² = - 0.83). This was attributed to South African samples being underrepresented in the global dataset. Sample state and sampling design were the most influential parameters affecting RMSE. To develop a national calibration algorithm, effort should be placed on developing accurate calibration algorithms for smaller areas that could be added to the national spectral library. Reliable SOCS maps are important for accurately assessing carbon credits. To address this gap, the study aimed to determine whether conventional mapping using ordinary kriging (OK) or digital soil mapping (DSM) with machine learning (ML) should be used for mapping SOCS at field scale in South Africa. Fifty samples from three depths, 0-5, 5-15 and 15-30 cm were collected from two farms in South Africa and analysed for SOC content using the TDC method, while for dry bulk density (ρb) a pedo-transfer was created for predicting the ρb at unsampled locations. Maps for SOC content, ρb and SOCS were created using OK and DSM with ML and assessed with evaluation statistics including root mean square error (RMSE) and Lin’s concordance correlation coefficient (ρc). A calculate-first approach and map-first approach were used for calculating the total SOCS for each study site. Findings of this study demonstrated that DSM with ML should be used rather than OK for mapping SOCS at field scale in South Africa. Utilising DSM with ML will lead to more accurate and reliable SOCS maps. However, the results also indicated that the SOC content, ρb and SOCS maps for DSM with ML need to be improved. To increase the accuracy of maps generated using DSM with ML, additional soil samples could be collected, and field scale covariates should be incorporated. The results were inconclusive regarding whether the calculate-first approach or the map-first approach should be used when utilising OK or DSM with ML. Although both approaches have several advantages and disadvantages, more research is needed to determine which approach should be used when utilising OK or DSM with ML. However, the study concluded that the calculate-first approach should be endorsed, due to this approach being simpler and might lead to more reliable and accurate SOCS estimations. There is an ongoing debate about the most appropriate approach for quantifying changes in soil organic carbon stocks SOCS. This study aimed to compare two SOCS quantification approaches, modelling and re-measure, on field scale in South Africa. The baseline assessment (2022) for two fields in South Africa was established beforehand and afterwards, fifty samples from three depth increments, 0-5, 5-15 and 15-30 cm were collected, and the SOCS calculated for two study sites (2024) using the re-measure approach and the modelling approach using the Rothamsted carbon (RothC) model. The sampling cost was also estimated against the value of sequestered SOCS (2022-2024) and compared between the two approaches. The main finding from the study indicated that it was inconclusive if a modelling or re-measure approach should be used for quantifying SOCS on field scale in South Africa. The spatial variation of SOC was the main factor influencing the total SOCS, sequestered SOCS and carbon credit values. The results indicated that the modelling approach using the RothC model slightly underestimated the total SOCS and sequestered SOCS when utilising all 50 observations for both study sites. The industry-represented, “One-observation”, did not have a consistent trend and showed overprediction and underprediction for both study sites using the different approaches. This had significant financial implications using both approaches. The study also found that the spatial variation of the SOC must be taken into account at the onset of measuring the SOC content in year one, independent of whether the SOCS would be modelled or re-measured. The carbon sequestration industry must take the model limitations (such as field scale dynamics and sensitivity to short-term changes) into account, and find ways, such as an improved sampling design to address these accuracies of RothC at field scale.Doctor of Philosoph